1. Field of the Invention
The present invention relates to the security of network devices within a computer network.
2. The Prior Art
As is well known, a computer network is formed of a plurality of network devices, such as computers, printers, file servers, etc., which are interconnected such that data communications may pass between the network devices. This functionality may typically be provided by using one or more communication hubs, each having a number of ports to each of which a network device may be connected. In the case where more than one hub is used, a link is provided from each of these to another also via ports on the hubs. In this fashion the network devices are interconnected in a star or tree topology with the communication hubs being the nodes in the arrangement.
Each network device in this topology is therefore connected to the network via a single data cable, which is connected to a port on one of the communication hubs which in turn provides access for that device to the rest of the network devices.
The physical items forming such a network may be numerous and may be spread over a wide area, for instance in an office environment. Also, network devices such as computers are typically quite high value items and thus susceptible to be stolen or otherwise removed without authority. It would however be impractical and perhaps prohibitively expensive to conduct physical checks on the network devices except in the smallest network installation.
For these reasons there have been previously proposed security or alarm systems which automatically monitor the presence of the network devices and provide a visual, audible or other indication upon the removal of a network device.
For instance, in U.S. Pat. No. 5,406,260 there is described a system and method for monitoring the connection of electronic equipment such as remote computer workstations, to a network via a communication link, and detecting the disconnection of the equipment from the network. This system includes current loops internally coupled to protected pieces of equipment and a low current power signal is provided to each of the current loops. Removal of a piece of protected equipment breaks the current flow through the associated current loop and an alarm can be activated.
In network equipment such as that designed for use in Ethernet 10 base T networks there is typically an isolated transformer provided on the data connection to the device, the primary coil of which may be used as the current loop which will be broken on removal of the network device.
In WO97/09667 a different approach to the monitoring of a computer system is described. In this arrangement, each of the monitored computers continuously collects information about itself and supplies the information to a separate alarm centre. The alarm centre activates an alarm if the collected information changes by a significant amount over time. This system thus relies on the presence of active local intelligence in the monitored computers.
WO96/29638 describes a further alarm system for network connected computer equipment. In this arrangement each workstation is provided with an alarm unit which can communicate alarm signals to an alarm centre to indicate an alarm condition.
In these latter two examples of the prior art it is necessary for the monitored equipment to be powered-up all the time during which monitoring takes place while in the first mentioned document above and in the latter two there is no detailed discussion of the control of the system in particular relating to how the system is armed. The whole network is monitored whenever the alarm system is in operation.